Type-1 diabetes (T1D) is hallmarked by autoimmune attack on the insulin-producing p cells of the pancreatic slets. Infiltration of the islets by autoimmune/inflammatory cells, or insulitis, occurs in a percentage of genetically susceptible individuals, and a fraction of these subjects progress to clinically overt diabetes. Our view of T1D pathogenesis, especially in humans, remains clouded and controversial in many regards. The main reasons for :his unsatisfying picture are the complexity and heterogeneity of T1D, and the fact that it is not usually diagnosed until very late, when most initiating and propagating autoimmune phenomena have already played out. One approach to surmounting the latter problem would be to develop a method to non-invasively image the pancreatic slets in vivo in real time, which has been one of the prime objectives of the collaborative JDC/MGH imaging program. The overall goals of this component of the program, Project 2, has been and remains first, to help evaluate the applicability of novel imaging methodologies (developed within Project 1 and the Core) to mouse models of T1D; second, to exploit successful visualization strategies to answer currently unresolved questions about disease pathogenesis in these models (certain in concert with Project 3); and third, to perform proof-of-principle experiments for translation of successful strategies to human T1D patients (under the auspices of Project 4). During the past funding cycle, Project 2 successfully developed a method to non-invasively image islet inflammation in murine T1D. This technique is based on visualizing, via magnetic resonance imaging (MRI) of magnetic nanoparticles (MNP), the microvascular changes (vessel dilation, vascular leakage) that routinely accompany inflammation, coupled with uptake by neighboring macrophages. A mock clinical trial on recent-onset diabetic mice clearly established the power of the MNP-MRI technique to distinguish between individuals that were or were not responding to anti-CD3 monoclonal antibody therapy. These promising results prompted initiation of a collaborative clinical trial to assess the ability of an analogous MNP, Combidex, to reflect vascular changes in association with human T1D (Project 4). For the next funding cycle, Project 2 proposes to apply the MNP-MRI technique to additional questions concerning the pathophysiology of T1D, but mostly to capitalize on exciting recent advances in fluorescent protein tomography (FTP) made by Project 1 and The Core. At the cutting-edge of whole-body live-rodent imaging, this technique permits non-invasive visualization of proteins and cells of interest in deep tissues (eg the pancreas) of living mice. The Specific Aims are: 1. To develop a system to co-image islet p-cell mass and p-cell activity non-invasively in vivo. A triple- transgenic mouse approach will be taken, exploiting the newly engineered red fluorescent proteins, mRaspberry and dKeima, in conjunction with FTP. Also under this Aim, a series of calibration and validation experiments will put the systems utility to the test. 2. To exploit the p-cell mass/activity co-imaging system to tackle outstanding issues concerning type-1 diabetes in mouse models. Questions to be tackled include: a) How does p-cell mass/activity change during the natural history of T1D in the NOD mouse model? b) How do genetic factors impinge on the mass/activity curves? c) Can the residual level of p-cell mass or activity predict onset of overt diabetes? d) Do different therapeutic regimens influence p-cell mass/activity? 3. To exploit MNP-MRI to guide dissection of the cellular and molecular elements predicting development of clinical diabetes in individual NOD mice. The novel capability of predicting if and approximately when a given mouse will convert from insulitis to diabetes will be used to address the influence of a NK cells b T helper cell subsets and c CD4+25+Foxp3+Treg cells within the infiltrate. Successful achievement of these Aims should provide novel insights into the pathogenesis of T1D, as well as furnish proof-of-principle for human imaging endeavors. Joslin Diabetes Center, Boston, MA PHS 398 (Rev. 04/06) PageM Form Page 2 Project 2: Imaging p-cell attack in mouse models of type-1 diabetes PI: MathiS, Diane PD: Mathis, Diane KEY PERSONNEL. See instructions. Use continuation pages as needed to provide the required information in the format shown below. Start with Principal Investigator(s). List all other key personnel in alphabetical order, last name first. Name eRA Commons User Name Organization Role on Project Mathis, Diane DMATHIS JDC PI OTHER SIGNIFICANT CONTRIBUTORS Name Organization Role on Project Kulkarni, Rohit JDC Consultant Weir, Gordon JDC Consultant Human Embryonic Stem Cells ^ No CH Yes If the proposed project involves human embryonic stem cells, list below the registration number of the specific cell line(s) from the following list: http://StemcellS.nih.gov/reqistrv/indeX.aSP. Usecontinuation pages as needed. If a specific line cannotbe referencedat this time, includea statement that one from the Registry will be used. Cell Line PHS 398 (Rev. 04/06) Page^O Form Page 2-continued Number the following pages consecutively throughout the application. Do not use suffixes such as 4a, 4b. Project 2: Imaging p-cell attack in mouse models of type-1 diabetes PI: Mathis, Diane PD: Mathis, Diane The name of the principal investigator/program director must be provided at the top of each printed page and each continuation page. PROJECT 2 TABLE OF CONTENTS Page Numbers Cover Page 93 Description,